"Cartel destabilization effect of leniency programs"

This paper investigates the theoretically and empirically unsettled question of the effect of the leniency programs on cartel duration, cartel fines and the length of the investigation. The fact that leniency programs were implemented in two different jurisdictions (EU and Spain) at different moments of time, and the exogeneity of the date of introduction, allow us to identify and quantify the effect of the programs on the outcomes using difference-in-difference program evaluation techniques. We empirically show that leniency programs destabilize existing cartels in the short run as expected from theory and previous empirical papers, and then dissuade the creation of new cartels in the long run. Deterrence effects dominate empirically in the long run, although theoretically they might not dominate, and previous empirical findings were inconclusive. Fines per firm increase substantially after the introduction of the leniency policy, despite whistleblowing firms are partially or totally exempted from fines. The duration of the investigation increases with the introduction of the leniency programs. Leniency programs have sharp and clear short-run cartel destabilization and long-run cartel dissuasion effects.